Clinical Orthopaedics and Related Research: December 2016 - Volume 474 - Issue 12 - p 2645–2654 doi: 10.1007/s11999-016-5084-9 Clinical Research

Estimating the Societal Benefits of THA After Accounting for Work Status and Productivity: A Markov Model Approach

Koenig, Lane, PhD1,a; Zhang, Qian, PhD2; Austin, Matthew, S., MD3; Demiralp, Berna, PhD1; Fehring, Thomas, K., MD4; Feng, Chaoling, PhD1; Mather, Richard, C., III, MD5; Nguyen, Jennifer, T., MPP1; Saavoss, Asha, BA1; Springer, Bryan, D., MD4; Yates, Adolph, J., Jr, MD6
Hip

Background Demand for total hip arthroplasty (THA) is high and expected to continue to grow during the next decade. Although much of this growth includes working-aged patients, cost-effectiveness studies on THA have not fully incorporated the productivity effects from surgery.

 

Questions/Purposes We asked: (1) What is the expected effect of THA on patients’ employment and earnings? (2) How does accounting for these effects influence the cost-effectiveness of THA relative to nonsurgical treatment?

 

Methods Taking a societal perspective, we used a Markov model to assess the overall cost-effectiveness of THA compared with nonsurgical treatment. We estimated direct medical costs using Medicare claims data and indirect costs (employment status and worker earnings) using regression models and nonparametric simulations. For direct costs, we estimated average spending 1 year before and after surgery. Spending estimates included physician and related services, hospital inpatient and outpatient care, and postacute care. For indirect costs, we estimated the relationship between functional status and productivity, using data from the National Health Interview Survey and regression analysis. Using regression coefficients and patient survey data, we ran a nonparametric simulation to estimate productivity (probability of working multiplied by earnings if working minus the value of missed work days) before and after THA. We used the Australian Orthopaedic Association National Joint Replacement Registry to obtain revision rates because it contained osteoarthritis-specific THA revision rates by age and gender, which were unavailable in other registry reports. Other model assumptions were extracted from a previously published cost-effectiveness analysis that included a comprehensive literature review. We incorporated all parameter estimates into Markov models to assess THA effects on quality-adjusted life years and lifetime costs. We conducted threshold and sensitivity analyses on direct costs, indirect costs, and revision rates to assess the robustness of our Markov model results.

 

Results Compared with nonsurgical treatments, THA increased average annual productivity of patients by USD 9503 (95% CI, USD 1446-USD 17,812). We found that THA increases average lifetime direct costs by USD 30,365, which were offset by USD 63,314 in lifetime savings from increased productivity. With net societal savings of USD 32,948 per patient, total lifetime societal savings were estimated at almost USD 10 billion from more than 300,000 THAs performed in the United States each year.

 

Conclusions Using a Markov model approach, we show that THA produces societal benefits that can offset the costs of THA. When comparing THA with other nonsurgical treatments, policymakers should consider the long-term benefits associated with increased productivity from surgery.

 

Level of Evidence Level III, economic and decision analysis.


Link to article